How to remove white spaces at the end of every value between delimiter in a file?-CodePudding

I have pipe delimited file that looks like this:

col1|col2|col3|col4
a   |b  | some value | hello world     
a   |b  | some value | hello world     
a   |b  | some value | hello world     
a   |b  | some value | hello world

The values in each column, including the last, have trailing white spaces. How do I remove trailing white spaces using sed so that my output looks like this?

col1|col2|col3|col4
a|b|some value|hello world
a|b|some value|hello world
a|b|some value|hello world
a|b|some value|hello world

This is what I have tried which removes spaces between the delimiters, but not at the end of last column:

sed -i -e "s/\s*|/|/g" somefile.txt

CodePudding user response：

You may use this sed:

sed -E 's/[[:blank:]]*(\|)[[:blank:]]*|[[:blank:]] $/\1/g' file

col1|col2|col3|col4
a|b|some value|hello world
a|b|some value|hello world
a|b|some value|hello world
a|b|some value|hello world

Here:

[[:blank:]] (\|)[[:blank:]]*: Match whitespaces before and after |
|: OR
[[:blank:]] $: Match whitespaces before end

Alternative solution in awk:

awk -F ' *\\| *' -v OFS='|' '{sub(/  $/, "", $NF)} 1' file

CodePudding user response：

You can use this sedcommand:

sed 's/[[:blank:]]*$//; s/[[:blank:]]*|[[:blank:]]*/|/g' file

If you want to remove leading blanks as well, then just add s/^[[:blank:]]*// to the sed expression above.

CodePudding user response：

1st solution: Here is one more awk variant, written and tested with your shown samples; written and tested in GNU awk.

awk -F'[[:space:]]*\\|[[:space:]]*' -v OFS="|" '
{
  sub(/[[:space:]] $/,"")
  $1=$1
}
1
'  Input_file

2nd solution: Making field separator in awk and using match function of GNU awk please try following.

awk -F '[[:space:]]*\\|[[:space:]]*' -v OFS='|' '
($1=$1) && match($0,/[[:space:]] $/){
  print substr($0,1,RSTART-1)
}
'  Input_file